A study on predicted in-hospital mortality in critically ill patients with coronary heart disease: analysis of the MIMIC-IV database

The objective of this research was to identify predictors that increase the likelihood of death during hospitalization in critically ill patients with coronary heart disease using a predictive model. We gathered clinical information from individuals diagnosed with coronary artery disease in the MIMIC-IV repository. Logistic regression analysis was used to examine the factors. The predictive nomogram was created using R software. Additionally, the model’s accuracy was evaluated. The clinical effectiveness was assessed through the utilization of clinical impact and decision curves. Out of 8261 patients in the study, 9.0% of those with coronary heart disease died in the hospital. Independent risk factors for mortality in ICU patients with coronary heart disease included age, gender, diabetes, MCV, RDW, ALB, prothrombin time, blood lactate levels, and mean blood glucose levels. The predictive factors were integrated into the model, and the nomogram was constructed using R software. The model achieved an area under the ROC curve of 0.761 with a 95% confidence interval of 0.740–0.781. Subsequently, a calibration curve was constructed, indicating a satisfactory level of fit. The model’s clinical effectiveness was evaluated using clinical impact curves and decision curves. The predictive model utilizing variables such as age, gender, diabetes status, MCV, RDW, ALB, prothrombin time, lactate, and average blood glucose levels demonstrates the ability to forecast the likelihood of in-hospital mortality among individuals diagnosed with coronary heart disease.